28 #include <queso/TKGroup.h>
42 template <
class V = GslVector,
class M = GslMatrix>
50 const std::vector<double> & scales,
97 void print(std::ostream & os)
const;
118 #endif // UQ_MALA_TK_H
A class representing a Gaussian vector RV.
const BayesianJointPdf< V, M > & m_targetPdf
const GaussianVectorRV< V, M > & rv(unsigned int stageId) const
Gaussian increment property to construct a transition kernel.
void setRVsWithZeroMean()
Sets the mean of the RVs to zero.
void updateLawCovMatrix(const M &covMatrix)
Scales the covariance matrix.
~MetropolisAdjustedLangevinTK()
Destructor.
bool setPreComputingPosition(const V &position, unsigned int stageId)
Sets the pre-computing positions m_preComputingPositions[stageId] with a new vector of size position...
virtual bool covMatrixIsDirty()
void clearPreComputingPositions()
Clears the pre-computing positions m_preComputingPositions[stageId].
This base class allows the representation of a transition kernel.
MetropolisAdjustedLangevinTK(const char *prefix, const BayesianJointPdf< V, M > &targetPdf, const std::vector< double > &scales, const M &covMatrix)
Default constructor.
bool symmetric() const
Whether or not the matrix is symmetric. Always 'true'.
virtual void cleanCovMatrix()
void print(std::ostream &os) const
TODO: Prints the transition kernel.
This class allows the representation of the MALA transition kernel with a scaled covariance matrix fo...
A class for handling Bayesian joint PDFs.